On Wed, 2012-03-21 at 10:08 -0300, Luis Claudio R. Goncalves wrote: > he examples there are from older kernels (probably 2.6.18 and 2.6.23). > Several changes happened since then, to the kernels and even to the general > model of the RT patch. Among these changes, more than once softirqs have > been folded inside ksoftirqd and later unfolded. As of now, they are folded > and it seems to have helped to minimize a performance loss someone (tglx?) > observed in the early 3.0.x-rt days. I discussed this in Chemnitz, and will be working to at least pull the timer softirq out into its own thread. We need to look at this again and see how we can fix it in a clean fashion. -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
