Re: [RFC PATCH] RDMA/srp: Fix use-after-free in srp_exit_cmd_priv

[Bart Van Assche]

On 6/24/22 16:47, Jason Gunthorpe wrote:
> On Fri, Jun 24, 2022 at 04:26:06PM -0700, Bart Van Assche wrote:
> > On 6/24/22 15:59, Jason Gunthorpe wrote:
> > > I don't even understand how get_device() prevents this call chain??
> > >
> > > It looks to me like the problem is srp_remove_one() is not waiting for
> > > or canceling some outstanding work.
> >
> > Hi Jason,
> >
> > My conclusions from the call traces in Li's email are as follows:
> > * scsi_host_dev_release() can get called after srp_remove_one().
> > * srp_exit_cmd_priv() uses the ib_device pointer. If srp_remove_one() is
> > called before srp_exit_cmd_priv() then a use-after-free is triggered.
>
> Shouldn't srp_remove_one() wait for the scsi_host_dev to complete
> destruction? Clearly it cannot continue to exist once the IB device
> has been removed

That sounds like an interesting approach to me. Li, do you perhaps want 
to implement this approach?

Thanks,

Bart.