On Thu, Nov 02, 2017 at 04:45:51PM +0000, David Howells wrote: > Peter Zijlstra <[email protected]> wrote: > > > > The moment atomic_dec_and_test() is called, *net is at liberty to > > > disappear, > > > so there's no way to find a waitqueue - unless that waitqueue is > > > guaranteed > > > not to be deallocated, eg. by being global. > > > > But any possible wait side will still need to dereference *net at an > > indeterminate point in the future to ascertain the value does now indeed > > read 0. > > Yeah, the assumption is that the *wait* side is where the deallocation is > happening. So I have something like: > > net->live = false; // Tell the cell managers to delete the cells > ... > foreach (cell) > queue_work(cell); > ... > wait_on_atomic_t(&net->cells_outstanding, atomic_t_wait, > TASK_UNINTERRUPTIBLE); > ... > destroy(net); > > > ... We can certainly start with your patch, as that does clean things up. > > Can I put that down as a Grudgingly-acked-by?
Yeah sure ;-)
