lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Luc Vanlerberghe (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (LUCENE-6409) LongBitSet.ensureCapacity overflows on numBits > Integer.MaxValue
Date Fri, 10 Apr 2015 14:14:12 GMT

    [ https://issues.apache.org/jira/browse/LUCENE-6409?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14489470#comment-14489470
] 

Luc Vanlerberghe edited comment on LUCENE-6409 at 4/10/15 2:13 PM:
-------------------------------------------------------------------

I added testBits2Words on TestLongBitSet with various values, concentrating on boundary cases

I found some more issues that may cause spurious side effects.
While trying to correct them, it turns out that the tests do 'illegal' things (like setting
bits outside the [0, numBits[ range by manipulating the backing array directly)
This should cause things like cardinality() to fail (because it doesn't use a mask for the
last long, probably for performance), but apparently that's never tested.

I propose that creating a new Long-/FixedBitSet based on an existing backing array should
assert (in debug mode) that all the 'ghost' bits with index >= numBits should be clear.

Perhaps I should create a new issue for this to allow the ensureCapacity bug to be closed
first.



was (Author: lvl):
I added testBits2Words on TestLongBitSet with various values, concentrating on boundary cases

I found some more issues that may cause spurious side effects.
While trying to correct them, it turns out that the tests do 'illegal' things causing (like
setting bits outside the [0, numBits[ range by manipulating the backing array directly)
This should cause things like cardinality() to fail (because it doesn't use a mask for the
last long, probably for performance), but apparently that's never tested.

I propose that creating a new Long-/FixedBitSet based on an existing backing array should
assert (in debug mode) that all the 'ghost' bits with index >= numBits should be clear.

Perhaps I should create a new issue for this to allow the ensureCapacity bug to be closed
first.


>  LongBitSet.ensureCapacity overflows on numBits > Integer.MaxValue 
> -------------------------------------------------------------------
>
>                 Key: LUCENE-6409
>                 URL: https://issues.apache.org/jira/browse/LUCENE-6409
>             Project: Lucene - Core
>          Issue Type: Bug
>          Components: core/other
>            Reporter: Luc Vanlerberghe
>
> LongBitSet.ensureCapacity calculates the number of longs required to store the number
of bits correctly and allocates a long[] accordingly, but then shifts the array length (which
is an int!) left by 6 bits.  The int should be cast to long *before* performing the shift.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message