<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://bugs.webkit.org/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.4.1"
          urlbase="https://bugs.webkit.org/"
          
          maintainer="admin@webkit.org"
>

    <bug>
          <bug_id>95748</bug_id>
          
          <creation_ts>2012-09-04 07:50:15 -0700</creation_ts>
          <short_desc>TestWebKitAPI&apos;s WTF.Lrint has four failing asserts on Android</short_desc>
          <delta_ts>2012-09-12 03:05:55 -0700</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>WebKit</product>
          <component>Tools / Tests</component>
          <version>528+ (Nightly build)</version>
          <rep_platform>Unspecified</rep_platform>
          <op_sys>Unspecified</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords></keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Peter Beverloo">peter</reporter>
          <assigned_to name="Peter Beverloo">peter</assigned_to>
          <cc>d-r</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>711117</commentid>
    <comment_count>0</comment_count>
    <who name="Peter Beverloo">peter</who>
    <bug_when>2012-09-04 07:50:15 -0700</bug_when>
    <thetext>Revision 127001 introduced the WTF.Lrint tests, which seem to have a 32/64bit mismatch when run on Android. Both sizeof(long int) and sizeof(long) yield 32 and it seems to default to the NDK&apos;s lrint().

I don&apos;t have time to follow up immediately, but will eventually get around to doing it. Dominik, since these tests aren&apos;t being run on any bots yet it&apos;s not reasonable to ask you to take a look, but if you have any immediate ideas that&apos;d be most welcome of course :-).


E/chromium(13749): [ RUN      ] WTF.Lrint
E/chromium(13749): *** Failure in ../../Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp:44
E/chromium(13749): Value of: pow(2.0, 52)
E/chromium(13749):   Actual: 4.5036e+15
E/chromium(13749): Expected: lrint(pow(2.0, 52) - 0.5)
E/chromium(13749): Which is: 2147483647
E/chromium(13749): *** Failure in ../../Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp:45
E/chromium(13749): Value of: pow(2.0, 52) - 2
E/chromium(13749):   Actual: 4.5036e+15
E/chromium(13749): Expected: lrint(pow(2.0, 52) - 1.5)
E/chromium(13749): Which is: 2147483647
E/chromium(13749): *** Failure in ../../Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp:47
E/chromium(13749): Value of: -pow(2.0, 52)
E/chromium(13749):   Actual: -4.5036e+15
E/chromium(13749): Expected: lrint(-pow(2.0, 52) + 0.5)
E/chromium(13749): Which is: -2147483648
E/chromium(13749): *** Failure in ../../Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp:48
E/chromium(13749): Value of: -pow(2.0, 52) + 2
E/chromium(13749):   Actual: -4.5036e+15
E/chromium(13749): Expected: lrint(-pow(2.0, 52) + 1.5)
E/chromium(13749): Which is: -2147483648
E/chromium(13749): [  FAILED  ] WTF.Lrint</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>718131</commentid>
    <comment_count>1</comment_count>
    <who name="Peter Beverloo">peter</who>
    <bug_when>2012-09-12 03:05:55 -0700</bug_when>
    <thetext>This was fixed in another patch by adding a sizeof(long int)==8 check around these asserts.</thetext>
  </long_desc>
      
      

    </bug>

</bugzilla>