On 30/04/2016 23:57, Andy Lutomirski wrote: >> > Should we kill __pvclock_read_cycles in favor of vread_pvclock? It looks >> > doable at a quick scan... >> > > The in-kernel version might have to be a bit different because it > needs to handle the !stable case. If !stable, it should just use the > current CPU's copy which means that, realistically, it should just > get_cpu and use the local copy unconditionally. Other than that, it > could look a lot like the vread_pvclock variant. > > But I agree, the current thing is incomprehensible.
It also lacks smp_rmb()s. One is more or less implicit in rdtsc, but you need one to separate __pvclock_read_cycles's reads of src->foo from pvclock_read_flags's read of src->version. Minfei, would you like to take a look? Paolo
