I setup a test where I have one git repo with two jobs pointing to two different folders. They pull each 1 minute, and their job do nothing, and succeed.

I end-up to get the same change trigger two builds for the same job.

Extract:
Job #5:
Started on Jan 17, 2012 5:45:49 AM
Using strategy: Default
[poll] Last Build : #3
[poll] Last Built Revision: Revision 88d40a9c206fe9d9d419a7c41a5c1eb25643fd3e (origin/solrIntegration)
Fetching changes from the remote Git repositories
Fetching upstream changes from http://oxgslcrsa01.oaifield.onasgn.com:8082/git/ops
Polling for changes in
Done. Took 0.6 sec
Changes found

Job #4:
Started on Jan 17, 2012 5:44:49 AM
Using strategy: Default
[poll] Last Build : #3
[poll] Last Built Revision: Revision 88d40a9c206fe9d9d419a7c41a5c1eb25643fd3e (origin/solrIntegration)
Fetching changes from the remote Git repositories
Fetching upstream changes from http://oxgslcrsa01.oaifield.onasgn.com:8082/git/ops
Polling for changes in
Done. Took 0.57 sec
Changes found

Expectations is that no build will ever be duplicated due to the same changeset. Also by default, a the SCM trigger should not trigger a new build if the same build have already been trigger manually.