Re: [PATCH 06/18] x86, barrier: stop speculation for failed access_ok

[Peter Zijlstra]

On Sun, Jan 07, 2018 at 06:57:35PM -0800, Alexei Starovoitov wrote:
> On Sun, Jan 07, 2018 at 01:59:35PM +0000, Alan Cox wrote:

> > lfence timing is also heavily dependent upon what work has to be done to
> > retire previous live instructions. 
> > BPF does not normally do a lot of writing so you'd expect the cost to be low.
> 
> right. to retire previous loads. Not sure what 'not a lot of writing'
> has to do with lfence.

LFENCE will wait for completion on _ALL_ prior instructions, not just
loads.

Stores are by far the most expensive instructions to wait for, as they
only complete once their value is globally visible (on x86).