Perfromance issues

[Chris Wright]

* Steve Grubb (sgrubb at redhat.com) wrote:
> On Thursday 08 September 2005 13:28, Chris Wright wrote:
> > Any micro optimisation needs to be benchmarked to show whether it's really
> > useful (I'd expect this one to be in the noise even though assembly
> > inspection should show improvement).
> 
> I think we have to start thinking about this. People are comparing system 
> performance with audit disabled and enabled and not compiled in at all. The 
> hit is significant. We need to see what can be done to make it better.

Absolutely agreed.  Profiles would be quite helpful here.

> > IMO, a better choice would be a static inline stub to keep from cluttering
> > call sites.  Smth. like: 
> 
> I like this. Easy to contain the changes without a big change to any patch. 
> But just to be sure, I'm not talking only about the one case I posted. 
> syscall_entry & exit are a bigger hit as it affects all syscalls.

Yup, generic change that potentially improves every call site.

thanks,
-chris