Bug 66956 – REGRESSION(r93755): It made 14 jsc test and ~500 layout test fail on Qt-ARM bot

• Status: RESOLVED
• Resolution: FIXED
• Component: JavaScriptCore

[Csaba Osztrogonác]

Check the results for details:
http://build.webkit.sed.hu/builders/ARMv5%20Linux%20Qt%20Release%20%28Test%29/builds/3203

[Oliver Hunt]

Could you include sample failures + stacktraces in these bugs?  It's much more useful than a slow loading link, and the given link doesn't include any stacktraces at all thus making it completely useless.

I suspect that the cause of these failures would be obvious to any Qt engineer who looked at them

[Csaba Osztrogonác]

Zoltan (our ARM expert) said that he'll check this bug tomorrow morning.

[Oliver Hunt]

(In reply to comment #2)
> Zoltan (our ARM expert) said that he'll check this bug tomorrow morning.


This should be trivial -- the problem is that the Qt bot isn't providing console output.  The problem is almost certainly going to be that the size of one of the getById insn sequences has changed and the constant just needs to be updated appropriately, the assertions for these branches even tell you what the new size should be.

[Gabor Loki]

Created attachment 105230 [details]
Rebaseline constants for patching GetByIdSlowCaseCall on ARM

A quick fix for r93755

[Oliver Hunt]

Comment on attachment 105230 [details]
Rebaseline constants for patching GetByIdSlowCaseCall on ARM

r=me!

[WebKit Review Bot]

Comment on attachment 105230 [details]
Rebaseline constants for patching GetByIdSlowCaseCall on ARM

Clearing flags on attachment: 105230

Committed r93832: <http://trac.webkit.org/changeset/93832>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.