On 2019-10-04 19:04:11 [+0200], Uladzislau Rezki wrote: > if another, we can free the object allocated on previous step if > it already has it. If another CPU does not have it, save it in > ne_fit_preload_node for another current CPU to reuse later. Further > we can not migrate because of: > > <snip> > spin_lock(&vmap_area_lock); > preempt_enable(); > <snip>
ah right. So you keep the lock later on, I somehow missed that part. That way it actually makes sense. Sebastian
