Another Software Patent That Should Be Tossed Out
Written by Michael Larabel in Free Software on 3 March 2015 at 08:52 AM EST. 16 Comments
Hewlett Packard has patented the identification of a failed code change by automatically running a set of tests against the set of changes for indicating the regression.

The abstract of WO 2013115797 A1 reads, "A method to identify a failed code change in a deployment pipeline with a plurality of code changes. The plurality of code changes are tested by running a set of tests on the plurality of code changes until a subset of the plurality of code changes pass the set of tests. Each time the subset fails the set of tests, at least one of the plurality of code changes is removed from the subset. The failed code change is determined based on the subset that passes the set of tests." The patent was filed in 2012 and published in 2013.

In a manual mode, it basically is like Git's bisecting abilities. This patent was brought to my attention via Twitter as there's individuals seeking prior art for this patent about automatically detecting a failed code change.

This was brought to my attention since a Phoronix reader had recalled I basically did this years prior to HP Software's patent ever being filed. Using git-bisect underneath on the Linux kernel or any other code-base, the Phoronix Test Suite as the "management tool" can automatically bisect a given code-base by comparing the test results before and after each commit in the bisect process compared to the good/bad results.

Back in 2009 I wrote about this automated bisecting to find kernel regressions via Git and the open-source Phoronix Test Suite. The code to it has been around since the end of 2009 and has been through several forms and revisions well before 2012 when the patent was filed. I've also written about it several times about using this method to automatically find various performance regressions in upstream code-bases. Outside of the Phoronix Test Suite, there's probably been others doing a similar process for automating the bisecting to find issues, which even last decade I wouldn't have considered a patent-able idea.

While looking through that silly patent after pointed out on Twitter, there's also a few other testing-related ones filed by HP at the same time. There's also WO 2013115786 A1 filed in early 2012 as "Continuous deployment of code changes". The abstract reads, "A processor implemented method to deploy a code change in a software application. The code change is assigned to a deployment pipeline based on a filtering rule. The code change is deployed after the code change passes a set of test criteria associated with the deployment pipeline."

This second patent sounds like various continuous integration software that's been out before. That too is another area I've explored in the past too with efforts for matters like making/verifying software/driver optimizations easier, etc. There should also be a lot of other prior art from the many CI software packages out there.
Related News
About The Author
Author picture

Michael Larabel is the principal author of and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and automated benchmarking software. He can be followed via Twitter or contacted via

Popular News This Week