On 2014/3/20 21:49, Viresh Kumar wrote:
For networking applications platforms need to provide one CPU per each user space data plane thread. These CPUs should not be interrupted by kernel at all unless userspace has requested for some syscalls. Currently, there are background kernel activities that are running on almost every CPU, like: timers/hrtimers/watchdogs/etc, and these are required to be migrated to other CPUs.
To achieve that, this patch adds another option to cpusets, i.e. 'quiesce'. Writing '1' on this file would migrate these unbound/unpinned timers/workqueues away from the CPUs of the cpuset in question. Writing '0' has no effect and this file can't be read from userspace as we aren't maintaining a state here.
This doesn't look like a complete solution, because newer timers/workqueues can still run in those CPUs. Seems like the proposal discussed is to support setting cpu affinity for workqueues through sysfs. If so, we can migrate workqueues when affinity is set, so we don't need this cpuset.quiesce ?
Currently, only timers are migrated. This would be followed by other kernel infrastructure later.
Suggested-by: Peter Zijlstra peterz@infradead.org Signed-off-by: Viresh Kumar viresh.kumar@linaro.org