JDK-6647451 : java/awt/font/Rotate/TextTransformTest.java failed
  • Type: Bug
  • Component: client-libs
  • Sub-Component: 2d
  • Affected Version: 6u10
  • Priority: P3
  • Status: Closed
  • Resolution: Fixed
  • OS: generic
  • CPU: generic,x86
  • Submitted: 2008-01-05
  • Updated: 2011-03-08
  • Resolved: 2011-03-08
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
6u10Fixed 7 b120Fixed
Related Reports
Duplicate :  
Relates :  
- java/awt/font/Rotate/TextTransformTest.java started to failed in jdk 6u10 b06
- test passed in 6u10 b05
- test passed in 6u4 b12
- test passed in 6u5 b10

command: main TextTransformTest
reason: User specified action: run main TextTransformTest 
elapsed time (seconds): 2.525
java.lang.RuntimeException: Test failed: hintId=1 strId=0 rotation=0 isExact=false
	at TextTransformTest.main(TextTransformTest.java:167)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
	at java.lang.reflect.Method.invoke(Method.java:597)
	at com.sun.javatest.regtest.MainWrapper$MainThread.run(MainWrapper.java:78)
	at java.lang.Thread.run(Thread.java:619)

EVALUATION First, difference is only present in the greyscale mode, LCD modes are ok. Reverting part of change for 6223022 (just rounding done in the t2k/util.c) seems to return old behavior. That rounding change helped to improve quality of with 11pt MS Sans Serif Q, 12pt Z and several other glyphs from other fonts/sizes. Therefore in general it is clearly positive and we do not want to rollback it. Note that difference in AA mode is also very small. It is because same fixed point multiplication operation is used to generate stems for AA and new rounding introduces small difference there. I am not sure what is the best way to approach this - we can try to revert to old rounding for AA calculations or inspect AA calculations and try to update them to not depend much on rounding. In any case this problem is not very significant. And improvement done by 6223020 justifies these sideeffects. Perhaps we can remove AA mode from regression test for now.

EVALUATION The only related fix in b05 was fix for 6223022. However, i still do not clearly understand why it could have caused this (difference is present in AA and LCD modes).