Fix for HADOOP-5079 HashFunction inadvertently destroys some randomness
-----------------------------------------------------------------------
Key: HADOOP-5255
URL: https://issues.apache.org/jira/browse/HADOOP-5255
Project: Hadoop Core
Issue Type: Bug
Components: io
Affects Versions: 0.20.0, 0.21.0
Reporter: stack
Priority: Minor
Fix For: 0.20.0, 0.21.0
HADOOP-5079 did this "HashFunction.hash restricts initval for the next hash to
the [0, maxValue) range of the hash indexes returned. This is suboptimal,
particularly for larger nbHash and smaller maxValue. Rather we should first set
initval, then restrict the range for the result assignment." The patch
committed on that issue introduced a new bug: "My first patch contained a
regression: you have to take the remainder before calling Math.abs, since
Math.abs(Integer.MIN_VALUE) == Integer.MIN_VALUE still" (Jonathan Ellis).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
