Re: [PATCH v7 3.2-rc2 5/30] uprobes: copy of the originalinstruction.

[Peter Zijlstra]

On Wed, 2011-11-23 at 14:49 -0500, Steven Rostedt wrote:
> On Wed, 2011-11-23 at 19:40 +0100, Peter Zijlstra wrote:
> > On Fri, 2011-11-18 at 16:37 +0530, Srikar Dronamraju wrote:
> > > +               /* TODO : Analysis and verification of instruction */
> > 
> > As in refuse to set a breakpoint on an instruction we can't deal with?
> > 
> > Do we care? The worst case we'll crash the program, but if we're allowed
> > setting uprobes we already have enough privileges to do that anyway,
> > right?
> 
> Well, I wouldn't be happy if I was running a server, and needed to
> analyze something it was doing, and because I screwed up the location of
> my probe, I crash the server, made lots of people unhappy and lose my
> job over it.
> 
> I think we do care, but it can be a TODO item.

But but but, why not let userspace sort it? And if you're going to
provide the kernel with inode:offset data yourself, you're already well
aware of wtf you're doing.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/