[
https://bro-tracker.atlassian.net/browse/BIT-1057?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13510#comment-13510
]
Matthias Vallentin commented on BIT-1057:
-----------------------------------------
Yup, this looks good to me.
For the record: the problem was indeed that the platform-specific block size of
the bit vector. This could have resulted in different hash values, because the
hashing algorithm took the underlying representations (i.e., the bit vector
blocks) as opposed to the individual bits themselves. Moreover, the {{fmt}}
function casted the 32-bit value into 64-bit one, potentially causing undefined
behavior.
I will merge your changes into {{topic/matthias/bloom-filter}} and only do
cosmetic polishing, such as using Bro's {{uint64}} instead of {{uint64_t}}.
> topic/jsiwek/bloomfilter-fix
> ----------------------------
>
> Key: BIT-1057
> URL: https://bro-tracker.atlassian.net/browse/BIT-1057
> Project: Bro Issue Tracker
> Issue Type: Patch
> Components: Bro
> Affects Versions: git/master
> Reporter: Jon Siwek
> Assignee: Matthias Vallentin
> Fix For: 2.2
>
>
> The change (1) I did on this branch makes the bloomfilter-seed test start
> passing on 32-bit systems, though I'm not confident how correct it is.
> Matthias, can you review it and change this to a merge request if it looks
> sane?
> (1) https://github.com/bro/bro/commit/774dadfe9aedc9fed98df69abcc83a3068bdf06b
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://bro-tracker.atlassian.net/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
_______________________________________________
bro-dev mailing list
[email protected]
http://mailman.icsi.berkeley.edu/mailman/listinfo/bro-dev
