Linux 3.6 Kernel Regression Causes ~20% Drop

[phoronix]

Phoronix: Linux 3.6 Kernel Regression Causes ~20% Drop

A commit within the in-development Linux 3.6 kernel has caused the PostgreSQL database server workload to regress by 15~20%. Fortunately, the commit has been spotted...

http://www.phoronix.com/vr.php?view=MTE4NTg

[Eisnefaust]

Well, such things happen all the day in a "in-development" system and therefore people/developers are happy having some laboratory rats testing. But is this worth a whole message on a news platform?

[DanL]

But is this worth a whole message on a news platform?

Benchmarking for regressions is important to the one guy who runs this site (and is interesting to at least some of its readers).

I'm really tired of people questioning the newsworthiness of every news blurb...

[russofris]

It's neat to compare kernel schedulers to what we have in nature. The CFS is a good, yet simple, demonstration of how reciprocity works. I wonder what the result would be if we decentralized the scheduler and made each thread schedule itself. Would an empathetic scheduler, where threads can cooperate together, work any better? Perhaps I should look at Frans De Waal's research and make a tech demo.

F