Re: [PATCH v3 3/3] IB/core: Obtain subnet_prefix from cache in IB devices.

[Leon Romanovsky]

On Wed, Jun 16, 2021 at 11:20:01AM +0000, Haakon Bugge wrote:
> 
> 
> > On 15 Jun 2021, at 18:13, Haakon Bugge <haakon.bugge@xxxxxxxxxx> wrote:
> > 
> > 
> > 
> >> On 15 Jun 2021, at 07:08, Leon Romanovsky <leon@xxxxxxxxxx> wrote:
> >> 
> >> On Mon, Jun 14, 2021 at 04:29:09PM +0000, Haakon Bugge wrote:
> >>> 
> >>> 
> >>>> On 14 Jun 2021, at 09:25, Leon Romanovsky <leon@xxxxxxxxxx> wrote:
> >>>> 
> >>>> On Mon, Jun 14, 2021 at 03:32:39AM +0000, Haakon Bugge wrote:
> >>>>> 
> >>>>> 
> >>>>>> On 9 Jun 2021, at 12:40, Leon Romanovsky <leon@xxxxxxxxxx> wrote:
> >>>>>> 
> >>>>>> On Wed, Jun 09, 2021 at 09:26:03AM +0000, Anand Khoje wrote:
> >>>>>>> Hi Leon,
> >>>>>> 
> >>>>>> Please don't do top-posting.
> >>>>>> 
> >>>>>> 
> >>>>>>> 
> >>>>>>> The set_bit()/clear_bit() and enum ib_port_data_flags  has been added as a device that can be used for future enhancements. 
> >>>>>>> Also, usage of set_bit()/clear_bit() ensures the operations on this bit is atomic.
> >>>>>> 
> >>>>>> The bitfield variables are better suit this use case.
> >>>>>> Let's don't overcomplicate code without the reason.
> >>>>> 
> >>>>> The problem is always that people tend to build on what's in there. For example, look at the bitfields in rdma_id_private, tos_set,  timeout_set, and min_rnr_timer_set.
> >>>>> 
> >>>>> What do you think will happen when, let's say, rdma_set_service_type() and rdma_set_ack_timeout() are called in close proximity in time? There is no locking, and the RMW will fail intermittently.
> >>>> 
> >>>> We are talking about device initialization flow that shouldn't be
> >>>> performed in parallel to another initialization of same device, so the
> >>>> comparison to rdma-cm is not valid here.
> >>> 
> >>> I can agree to that. And it is probably not worthwhile to fix the bit-fields in rdma_id_private?
> >> 
> >> Before this article [1], I would say no, we don't need to fix.
> >> Now, I'm not sure about that.
> >> 
> >> "He also notes that even though the design flaws are difficult to exploit
> >> on their own, they can be combined with the other flaws found to make for
> >> a much more serious problem."
> >> 
> >> and 
> >> 
> >> "In other words, people did notice this vulnerability and a defense was standardized,
> >> but in practice the defense was never adopted. This is a good example that security
> >> defenses must be adopted before attacks become practical."
> > 
> > Let me send you a commit tomorrow. The last sentence you quoted above is ambiguous as far as I can understand. But the intention is clear though :-)
> 
> Do you prefer for-next or for-rc for this?

for-next, please.

Thanks

> 
> Thxs, Håkon
>