On Wed, Jun 06, 2018 at 02:05:39PM +0200, Andrea Parri wrote: > Hi Daniel, Viresh, > > On Wed, Jun 06, 2018 at 04:15:28PM +0530, Viresh Kumar wrote: > > On 06-06-18, 12:22, Daniel Lezcano wrote: > > > (mb() are done in the atomic operations AFAICT). > > To do my bit, not all atomic ops do/imply memory barriers; e.g., > > [from Documentation/atomic_t.txt] > > - non-RMW operations [e.g., atomic_set()] are unordered > > - RMW operations that have no return value [e.g., atomic_inc()] are > unordered
Quite so indeed. > > AFAIU, it is required to make sure the operations are seen in a particular > > order > > on another CPU and the compiler doesn't reorganize code to optimize it. > > > > For example, in our case what if the compiler reorganizes the atomic-set > > operation after wakeup-process ? But maybe that wouldn't happen across > > function > > calls and we should be safe then. > > IIUC, wake_up_process() implies a full memory barrier and a compiler barrier, > due to: Yes, the wakeup being a RELEASE (at least) is a fairly fundamental property for causality. You expect the woken task to observe the condition it got woken up on.