Re: [PATCH 00/30] C/R OpenVZ/Virtuozzo style

[Alexey Dobriyan <adobriyan@gmail.com>]

On Tue, Apr 14, 2009 at 03:44:20PM +0200, Ingo Molnar wrote:
> 
> * Alexey Dobriyan <adobriyan@gmail.com> wrote:
> 
> > On Mon, Apr 13, 2009 at 11:39:51AM -0700, Linus Torvalds wrote:
> > > 
> > > 
> > > On Mon, 13 Apr 2009, Alexey Dobriyan wrote:
> > > > 
> > > > Well, in OpenVZ everything is in kernel/cpt/ and prefixed with "cpt_"
> > > > and "rst_".
> > > 
> > > So?
> > > 
> > > We're not merging OpenVZ code _either_.
> > 
> > This is to give example of other prefixes: cpt_ and rst_
> > Are they fine?
> 
> Not really. 'rst' can be easily mistaken for 'reset' and neither 
> really tells me at a glance what they do. They are also quite 
> tongue-twisters.
> 
> See my namespace analysis and suggestions from yesterday for a 
> proper naming scheme.
> 
> The key i believe is to move away from this singular 'the world is 
> all about checkpoint and restore', and move it to a IMHO clearer 
> state_*() type of naming which really isolates all these kernel 
> state save/restore management APIs from other kernel APIs. (See my 
> mail from yesterday for details.)
> 
> kstate_*() would be another, perhaps even clearer naming scheme. 
> I.e.:
> 
>   kstate_checkpoint_XYZ()
>   kstate_restore_XYZ()
>   kstate_collect_XYZ()
>   kstate_dump_XYZ()
>   kstate_image_XYZ()
>   ...
> 
> Just _look_ at them - they are expressive at a glance, and 
> reasonably short. That is the kind of first-time impression
> we need, not a 'wtf?' moment.
> 
> I just checked, there's zero hits on "git grep \<kstate_" in the 
> kernel, so it's a pristine namespace. IMHO, go wild ...

Need to try it for real.

One minor nit. This kstate_ doesn't include quite a noticable of
in-kernel state. For example, task readahead state isn't relevant
to C/R at all. It's state but irrelevant state.