Re: [PATCH v5] sched/fair: Fix unfairness caused by stalled tg_load_avg_contrib when the last task migrates out.

[Aaron Lu]

On Fri, Nov 28, 2025 at 02:40:17PM +0100, Peter Zijlstra wrote:
> On Fri, Nov 28, 2025 at 07:54:45PM +0800, Aaron Lu wrote:
> > Hello,
> > 
> > On Wed, Aug 27, 2025 at 10:22:07AM +0800, xupengbo wrote:
> > > When a task is migrated out, there is a probability that the tg->load_avg
> > > value will become abnormal. The reason is as follows.
> > > 
> > > 1. Due to the 1ms update period limitation in update_tg_load_avg(), there
> > > is a possibility that the reduced load_avg is not updated to tg->load_avg
> > > when a task migrates out.
> > > 2. Even though __update_blocked_fair() traverses the leaf_cfs_rq_list and
> > > calls update_tg_load_avg() for cfs_rqs that are not fully decayed, the key
> > > function cfs_rq_is_decayed() does not check whether
> > > cfs->tg_load_avg_contrib is null. Consequently, in some cases,
> > > __update_blocked_fair() removes cfs_rqs whose avg.load_avg has not been
> > > updated to tg->load_avg.
> > > 
> > > Add a check of cfs_rq->tg_load_avg_contrib in cfs_rq_is_decayed(),
> > > which fixes the case (2.) mentioned above.
> > > 
> > > Fixes: 1528c661c24b ("sched/fair: Ratelimit update to tg->load_avg")
> > > Tested-by: Aaron Lu <ziqianlu@xxxxxxxxxxxxx>
> > > Reviewed-by: Aaron Lu <ziqianlu@xxxxxxxxxxxxx>
> > > Reviewed-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> > > Signed-off-by: xupengbo <xupengbo@xxxxxxxx>
> > 
> > I wonder if there are any more concerns about this patch? If no, I hope
> > this fix can be merged. It's a rare case but it does happen for some
> > specific setup.
> > 
> > Sorry if this is a bad timing, but I just hit an oncall where this exact
> > problem occurred so I suppose it's worth a ping :)
> 
> Totally missed it. Seems okay, let me go queue the thing.

Thanks Peter!