Re: Nick's core remove PageReserved broke vmware...

[Hugh Dickins <hugh@veritas.com>]

On Thu, 10 Nov 2005, Michael S. Tsirkin wrote:
> Quoting Hugh Dickins <hugh@veritas.com>:
> > 
> > You're right, and it would be a good choice, except that MAP_INHERIT on
> > some OSes has a particular meaning (about inheriting across an exec),
> > so I think avoid confusion with that.  MADV_DONTFORK and MADV_DOFORK?
> > Accompanied by VM_DONTFORK?
> 
> Its actually similiar.

Indeed, similar, but different.

> Maybe MADV_FORK_INHERIT/MADV_FORK_DONT_INHERIT then?

Those names are a lot longer than the others in that file.
I still prefer MADV_DONTFORK etc. myself.  Or the original MADV_DONTCOPY,
though I think you were quite right to point out it's all about forking.

> I find using "COPY" there confusing, since the copy is only done on write ...

There are lots of levels on which there's copying or not.  The name
VM_DONTCOPY means "don't copy this vma when forking", it's not thinking
of copying pages or even of copying ptes: just don't copy this vma into
the child.

Hugh