[PATCH 1/1] sched/core: Fix stuck on completion for affine_move_task() when stopper disable

Peter Zijlstra peterz at infradead.org
Thu Sep 28 08:19:17 PDT 2023


On Thu, Sep 28, 2023 at 05:16:16PM +0200, Peter Zijlstra wrote:

> AFAICT this is migrate_enable(), which acts on current, so how can the
> CPU that current runs on go away?

> Your patch has at least one very obvious resource leak.

Sorry those are not so, I ended up staring at the wrong
stop_one_cpu_nowait() :-/

Still, the rest is very much the case, if you can't describe the exact
race scenario, you can't be talking about a solution.



More information about the Linux-mediatek mailing list