On Tue, Aug 11, 2020 at 09:26:37AM +0200, Oleg Nesterov wrote: > On 08/11, Oleg Nesterov wrote: > > > > On 08/11, Peter Zijlstra wrote: > > > > > > On Tue, Aug 11, 2020 at 08:45:16AM +0200, Oleg Nesterov wrote: > > > > > > > > ->jobctl is always modified with ->siglock held, do we really need > > > > WRITE_ONCE() ? > > > > > > In theory, yes. The compiler doesn't know about locks, it can tear > > > writes whenever it feels like it. > > > > Yes, but why does this matter? Could you spell please? > > Do you mean that compiler can temporary set/clear JOBCTL_TASK_WORK > when it sets/clears another bit? Possibly, afaict the compiler is allowed to 'spill' intermediate state into the variable. If any intermediate state has the bit clear,...
