Bug 201902 – REGRESSION: [iOS 13?] TestWebKitAPI.SharedBufferTest.tryCreateArrayBufferLargeSegments is failing

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Ryan Haddad]

The following test is failing on iOS simulator bots running iOS 13:

    TestWebKitAPI.SharedBufferTest.tryCreateArrayBufferLargeSegments
        
        /Volumes/Data/slave/ios-simulator-13-release/build/Tools/TestWebKitAPI/Tests/WebCore/SharedBuffer.cpp:117
        Expected equality of these values:
          'a'
            Which is: 'a' (97, 0x61)
          static_cast<char*>(arrayBuffer->data())[position]
            Which is: '\0'

https://build.webkit.org/builders/Apple%20iOS%2013%20Simulator%20Release%20WK2%20%28Tests%29/builds/2/steps/run-api-tests/logs/stdio

[Radar WebKit Bug Importer]

<rdar://problem/55466168>

[Ryan Haddad]

Disabled the test for iOS in https://trac.webkit.org/changeset/250105

[Alex Christensen]

Interesting.  The test passes for me locally.  Maybe it's misinterpreting the Vector constructor somehow.

[Alex Christensen]

Created attachment 380797 [details]
Patch

[Alex Christensen]

Comment on attachment 380797 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=380797&action=review

> Source/WTF/wtf/Vector.h:-208
> -        if (!__builtin_constant_p(dstEnd - dst) || (!(dstEnd - dst)))

This line should have been ... || (!!(dstEnd - dst)))
That was causing this to not call memset only in optimized builds when the size was nonzero.
Since I think it's not needed let's just remove it and let the compiler optimize out any memset with 0 size.

[Alex Christensen]

Ping review?

[Alex Christensen]

http://trac.webkit.org/r251089

[Jiewen Tan]

*** Bug 201620 has been marked as a duplicate of this bug. ***