Re: [PATCH] sched: Fix potential near-infinite distribute_cfs_runtime loop

2014-06-22 Thread Peter Zijlstra
On Fri, Jun 20, 2014 at 03:21:20PM -0700, Ben Segall wrote: > distribute_cfs_runtime intentionally only hands out enough runtime to > bring each cfs_rq to 1 ns of runtime, expecting the cfs_rqs to then take > the runtime they need only once they actually get to run. However, if > they get to run su

[PATCH] sched: Fix potential near-infinite distribute_cfs_runtime loop

2014-06-20 Thread Ben Segall
distribute_cfs_runtime intentionally only hands out enough runtime to bring each cfs_rq to 1 ns of runtime, expecting the cfs_rqs to then take the runtime they need only once they actually get to run. However, if they get to run sufficiently quickly, the period timer is still in distribute_cfs_runt