Re: [RFC] [PATCH 2.6.37-rc5-tip 5/20] 5: Uprobes: register/unregister probes.

[Srikar Dronamraju <srikar@linux.vnet.ibm.com>]

* Peter Zijlstra <peterz@infradead.org> [2011-01-27 11:23:37]:

> On Thu, 2011-01-27 at 15:31 +0530, Srikar Dronamraju wrote:
> > > > >  - validate that the vma is indeed a map of the right inode
> > > > 
> > > > We can add a check in write_opcode( we need to pass the inode to
> > > > write_opcode).
> > > 
> > > sure..
> > > 
> > > > >  - validate that the offset of the probe corresponds with the stored
> > > > > address
> > > > 
> > > > I am not clear on this. We would have derived the address from the
> > > > offset. So is that we check for
> > > >  (vaddr == vma->vm_start + uprobe->offset)
> > > 
> > > Sure, but the vma might have changed since you computed the offset -)
> > 
> > If the vma has changed then it would fail the 2nd validation i.e vma
> > corresponds to the uprobe inode right. If the vma was unmapped and
> > mapped back at the same place, then I guess we are okay to probe.
> 
> It can be unmapped and mapped back slightly different. A map of the same
> file doesn't need to mean its in the exact same location or has the
> exact same pgoffset.
> 
> 

If its not at the exact same location, then our third validation of
checking that (vaddr == vma->vm_start + uprobe->offset)  should fail
right?

Also should it be (vaddr == uprobe->offset + vma->vm_start -
vma->pgoff << PAGE_SHIFT) ?

-- 
Thanks and Regards
Srikar