Re: [PATCH v3] x86/uprobes: implement x86 specific arch_uprobe_*_step

[Ananth N Mavinakayanahalli <ananth@in.ibm.com>]

On Thu, Aug 30, 2012 at 04:37:24PM +0200, Oleg Nesterov wrote:
> On 08/30, Ananth N Mavinakayanahalli wrote:
> >
> > On Wed, Aug 29, 2012 at 07:37:48PM +0200, Oleg Nesterov wrote:
> > >
> > > Ananth, Sebastian, what if we start with the patch below? Then
> > > we can change arch/x86/kernel/uprobes.c to use the static
> > > uprobe_*_step() helpers from the 2nd patch.
> >
> > In principle I am fine with the change.
> 
> OK, good.
> 
> > > If we agree this code should be per-arch, then why do need other
> > > hooks? This is just ugly, we already have arch_pre/post_xol.
> > >
> > > The only problem is the pending powerpc patches, the change below
> > > obviously breaks them. Were they already applied? If not, then
> > > probably Ananth can do v6 on top of the patch below ;) The necessary
> > > fixup is trivial.
> >
> > They are under review.
> 
> OK, I understand that v6 can confuse the maintainer and complicate the
> merging process, please forget about v6.
> 
> And yes, this is really minor problem, still it would be nice to avoid
> the unnecessary hooks/complications...
> 
> So. We can add "weak arch_uprobe" hooks, fix x86, and after powerpc is
> merged change both powerpc and x86 in one patch (remove "weak" hooks
> and move enable/disable into arch_pre/post_xol).
> 
> Or. We can apply the patch I sent right now, you can fix powerpc later,
> when it is merged. This all is for 3.7 anyway, and fixup is trivial.
> 
> I agree either way. Which way do you prefer?

I prefer fixing both together later, just so nothing breaks while intial
testing, etc.

Ananth